The Unbelievable Insecurity of the Big Data Stack
Sheila A. Berta exposes critical security oversights across enterprise Big Data stacks (Hadoop, Spark, Flink), executing remote code execution on distributed analytics clusters.
Summary
The Growing Attack Surface of Distributed Analytics
Enterprise Big Data stacks combine complex distributed components—including Hadoop, Apache Spark, Flink, and Kafka—to ingest and process massive telemetry streams.
Architectural Complexity and Default Insecurity
Sheila A. Berta demonstrates that speed-first architectures and loose default access controls leave internal RPC endpoints and worker daemons exposed without authentication.
Chaining Flaws to Achieve Distributed Cluster RCE
By exploiting deserialization flaws, unauthenticated job submission APIs, and command injection gadgets, an attacker can compromise cluster nodes and exfiltrate raw enterprise data lakes.
Hardening Big Data Pipelines
Berta provides comprehensive security baselines, network segmentation rules, and authentication enforcement guides to protect distributed analytics infrastructure.
Related talks
Main Stage·45 min
ayoul3
ayoul3 details attacking Apache Spark big data clusters, demonstrating how misconfigurations and unauthenticated APIs yield root shells across thousands of nodes.
Main Stage·28 min
ayoul3
ayoul3 answers audience questions regarding Apache Spark cluster penetration testing, REST API security, and big data isolation.
Kaspersky·4 min
Insights on CyberSecurity Attacks
Presenters detail vulnerability discovery and exploitation in IT Security Challenges in Cloud, BYOD and Big Data Video | Insights on CyberSecurity Attacks.